Application of PCATS/DEBNetS to Regional Planning and Policy Analysis: Micro Simulation Studies in the Cities of Osaka and Kyoto, Japan
KITAMURA R, FUJII S, YAMAMOTO T and KIKUCHI A, Kyoto University, Japan
The problems and resulting limitations of conventional, trip-based passenger travel demand forecasting methods, such as the four-step procedures, have repeatedly been pointed out. For example, the absence of the dimension of time of day in these methods l
The problems and resulting limitations of conventional, trip-based passenger travel demand forecasting methods, such as the four-step procedures, have repeatedly been pointed out. For example, the absence of the dimension of time of day in these methods limits their applicability to the analysis of air pollution due to automotive traffic because the formation of photochemical smog depends on the clock time when NOÃ? and other substances are emitted.
Likewise, these methods cannot address how various demand management schemes may influence the frequency of cold starts. Being trip-based, these methods may not adequately capture how patterns of trip chaining may be influenced by policy measures, and how these changes may in turn affect traffic flow and pollutant emissions.
As an alternative to these conventional approaches, and as a more exact and versatile tool for demand forecasting and policy analysis, a micro-simulation model system of individuals' dally travel behavior is proposed in this study. The main body of the system comprises two simulators, PCATS and DEBNetS. PCATS stands for Prism Constrained Activity-Travel Simulator, and, as its name indicates, simulates an individual's daily activity and travel while explicitly incorporating H/igerstrand's space-time prism (H/igerstrand, 1970). DEBNetS, which stands for Dynamic Event-Based Network Simulator, is a micro-meso scale simulator of network flow. These two simulators are integrated to perform, in the nomenclature of the conventional four-step procedures, trip generation, trip distribution, modal split and network assignment.
There are several important differences between the conventional procedures and the proposed micro-simulation system that are expected to facilitate more authentic representation of activity-travel behavior by the proposed system. They are:
* Clock time is the key dimension of the proposed simulation system; activities and trips are generated for each person along a continuous time axis;
* The series of activities and trips made by an individual are simulated for the period of one day; thus spatial and temporal continuity of activities and travel is maintained and interdependencies among trip attributes are represented; and
* Hagerstrand's prisms are evaluated for each individual in the simulation along with coupling constraints associated with private modes of travel and operating hours of public transit; activities and trips are generated within the confines of these constraints.
A third component of the micro-simulation system is Household Attribute Generation System (HAGS), which populates each geographical zone with synthetic households while observing marginal distributions of pertinent variables in census and other data, then produce their future demographic and socio-economic attributes for long-range forecasting.
In this paper, PCATS, DEBNetS and HAGS are first described briefly, then two application examples of the proposed simulation system are presented. In the first case study, the model system is applied to evaluate the CO2 emissions reduction potential of a selected set of transportation control measures (TCMs) in the City of Kyoto, Japan. In the second case study, the system is applied to evaluate alternative long-range transportation plans in the City of Osaka, Japan. The objective of this paper is to demonstrate that the proposed micro- simulation approach is a viable planning tool that are based on a more holistic representation of urban residents' activity and travel.
Association for European Transport